On the Impact of Deep Learning and Feature Extraction for Arabic Audio Classification and Speaker Identification.
Sakib ShahriarRozita DaraKadhim HayawiPublished in: AICCSA (2022)
Keyphrases
- speaker identification
- feature extraction
- deep learning
- mel frequency cepstral coefficients
- speech recognition
- image classification
- pattern recognition
- feature vectors
- gaussian mixture model
- unsupervised learning
- feature selection
- machine learning
- feature set
- classification accuracy
- noisy environments
- dimensionality reduction
- broadcast news
- image processing
- support vector
- decision trees
- training set
- feature space
- speech signal
- principal component analysis
- supervised learning
- text mining
- denoising
- bag of words
- face recognition
- audio features
- active learning